Pneumococcal macrolide resistance—not a myth

Sir, In an article appearing in a recent issue of the Journal,1 Amsden attempted to ‘demythologize’ the significance of pneumococcal resistance to macrolide antibiotics. He suggested that patients with infections caused by strains of Streptococcus pneumoniae resistant to these antibiotics could be treated successfully with azithromycin and perhaps other macrolides. He regrets that ‘As a result of these in vitro reports, many prescribers are changing their choice of antibiotics for pneumococcal, as well as for community-acquired respiratory tract infections in general, away from macrolides.' I believe that the microbiological and in vivo data used by Amsden to support his proposal should be re-evaluated. With regard to the in vivo data, the clinical trials used as evidence of the efficacies of macrolides as therapy for patients with upper and lower respiratory tract infections do not in fact allow the effects of antibiotic resistance on the response to treatment to be determined. Furthermore, it is misleading to use the efficacy of azithromycin as therapy for patients with acute otitis media, a disease for which there is a high spontaneous cure rate, as evidence of the high activity of this drug against S. pneumoniae. Finally, the inclusion of studies that employed cefaclor, a β-lactam with poor in vitro activity against S. pneumoniae,2,3 as the comparator agent does little to strengthen the case for prescribing macrolides to patients with infections caused by strains shown to be resistant in vitro to these drugs. As described by Amsden, pneumococci possess, at different prevalences, two principal mechanisms of resistance to macrolide antibiotics. One involves target site modification by a methylase that is encoded by the ermB gene; such strains express the MLSB phenotype. The second involves an active efflux pump which removes only 14-membered and 15-membered macrolides from the cell; this mechanism is encoded by the mefE gene and strains harbouring this gene express the M phenotype. Strains expressing the MLSB phenotype exhibit cross-resistance, which is either constitutive or inducible, to macrolides, lincosamides and streptogramin B-type antibiotics. They exhibit high-level resistance to macrolides (MICs of erythromycin >32 mg/L) and are the most prevalent of the erythromycin-resistant (ErR) pneumococci, accounting for >90% of strains in Spain4 and 40–60% of strains in the USA and Canada.5 In an attempt to demonstrate the very high MICs of clarithromycin and azithromycin for these strains, I determined the in vitro susceptibilities of 53 recent clinical isolates of S. pneumoniae expressing the MLSB phenotype. I also assessed the effect on the MICs of the presence of CO2 in the incubation atmosphere, this being a variable of susceptibility testing to which Amsden attributes the falsely high MICs in some studies. The MICs of clarithromycin and azithromycin were determined by the microbroth dilution, agar dilution and Etest methods; the microbroth and agar dilution methods were performed according to the recommendations of the National Committee for Clinical Laboratory Standards6 and the Etest (AB Biodisk, Solna, Sweden) was performed according to the manufacturer's instructions. Cation-adjusted Mueller–Hinton broth or agar (Becton-Dickinson, Cockeysville, MD, USA) supplemented with 3–5% lysed horse blood was the medium used in all cases. For the microbroth dilution method the inoculum was 108 cfu/L and the strains were incubated in air, for the agar dilution method the inoculum was 104 cfu/spot and incubation was in air containing 5% CO2, and for the Etest the inoculum was 5 × 1010 cfu/L and duplicate plates were incubated in either air or air containing 5% CO2. S. pneumoniae ATCC 49619 was used as a control and the MICs were recorded after incubation at 35°C for 20–24 h. Though a small percentage of pneumococci harbouring the ermB gene exhibit low-level resistance to erythromycin, clarithromycin and azithromycin, the majority exhibit high to very high levels of resistance. In this study the MICs of the macrolides were lower when susceptibility testing by the Etest was performed in air, compared with an atmosphere containing CO2 (Table). None the less, the MIC90s of both clarithromycin and azithromycin for the ErR strains expressing the MLSB phenotype were >256 mg/L, irrespective of the method of determining the MICs and of whether or not CO2 was included in the incubation atmosphere (Table). These MICs are much higher than the concentrations of azithromycin measured at the site of infection and more than three-fold greater than the peak intracellular concentration (>80 mg/L) with which, according to Amsden, the pneumococci would be in contact within phagocytic cells. For infections caused by strains expressing the M phenotype (low-level resistance, i.e. MICs of erythromycin 2–16 mg/L), it may be appropriate to use microbiological or pharmacological criteria to speculate on the significance of the in vitro resistance of S. pneumoniae isolates to macrolides. However, even for those infections caused by strains exhibiting low-level resistance it will be necessary to provide in vivo evidence of clinical efficacy before changes in the present attitude to resistance will be adopted. For the time being, macrolides should not be used as therapy for patients with infections caused by pneumococci identified by in vitro susceptibility testing as resistant to these drugs. Table. MICs (mg/L) of clarithromycin and azithromycin for 53 clinical isolates of S. pneumoniae expressing the MLSB phenotype, as determined by three susceptibility testing methods Table. MICs (mg/L) of clarithromycin and azithromycin for 53 clinical isolates of S. pneumoniae expressing the MLSB phenotype, as determined by three susceptibility testing methods

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.